Australian (ASX) Stock Market Forum

Scalping Lynas experiment

Jump to new
Summary to date

Trade size = $50,000 (paper trading at present)
S1 = sell next day's open
[S2 = sell same day's close, purely for comparison, I expect this to work out better].


Mon___B:1.15___S1:1.16___S2:1.15
Tues___B:1.16___S1: 1.17___S2:1.15
Wed___B:1.17___S1: 1.14___S2:1.19
Thurs___B:1.145
 
The buys are real volume/time. No slippage.

I'm assuming no slippage at sell = next open or same day close because for a $50,000 amount, that shouldn't alter the price too much for LYC (historically, that is).
 
Mon___B:1.15___S1:1.16___S2:1.15
Tues___B:1.16___S1: 1.17___S2:1.15
Wed___B:1.17___S1: 1.14___S2:1.19
Thurs___B:1.145___S1:1.19___S2:1.17
Fri___B:1.19___
 
The problem GB

Is that when you do a look back on a test you can define the low as its already there.
In realtime it isnt---your taking an intelligent guess.
Youll be wrong far more often as you just wont know where the low is until close.
Your realtime results will vary dramatically from testing.
As the test knows where the low is.

As time goes by youll definately see what I mean.
 
tech after 5 days I'm well within the boundaries I set.

Mon - B = +.015 from low
Tues - B = +.01 from low
Wed - B = +.01 from low
Thurs - B = +.005 from low
Fri - B = likely+.015 ? wait and see.

--Using same day close = sell, profits are negligible at $90.
--Using next open = sell, profit is $2000, which is 4% return on capital in a week (depending on how Monday goes of course).

Yes, this is about intelligent guessing. Isn't everything?
 
Gringotts Bank said:
tech after 5 days I'm well within the boundaries I set.

Mon - B = +.015 from low
Tues - B = +.01 from low
Wed - B = +.01 from low
Thurs - B = +.005 from low
Fri - B = likely+.015 ? wait and see.

--Using same day close = sell, profits are negligible at $90.
--Using next open = sell, profit is $2000, which is 4% return on capital in a week (depending on how Monday goes of course).

Yes, this is about intelligent guessing. Isn't everything?
Click to expand...

Not a critisism just an observation.
One of the reasons people find their systems dont work in R/T like they do in testing.
But 5 days is hardly a statistically significant time period. I have a feeling (and time will tell) that one very bad day where the low turns much lower and the next days low gaps badly lower could wipe out all gains.

But so far so good and good on you for putting it up!

No not a guess but can be an anticipation.
A guess is a floppy point of maybe---this is it.
Anticipation is -----From here if this happens do this---if not do this.
 
I wouldn't call this a system, but a strategy based upon system backtesting (because of the need for discretion on entry).

--Entry is discretionary so long as it's <+.015 from LOW, then it will make you 100's of percent pa. Last year it was 300%.
--Exit is defined.

Even though next open was a better exit this week, historically, exiting the same close is far more profitable.

A large chunk of stocks will make you good money if you can buy near the day's low and sell at close. What made Lynas different to almost all other stocks is the sort of profit it returned if you got within 3 ticks of the low. This trend will definitely end when LYC either 1. morphs into a blue chip or 2. falls off the radar. But that point will be very obvious, and it won't hurt the bottom line.

Given that the day's range is often set by 11am, the screen time devoted to picking an entry is fairly minimal. If I was to trade it for real, I would definitely sell at close. I want to try my hand at a few days where the range is much higher. This week has been low in volatility. A trade size of $30,000-$50,000 is required to gain maximum % return benefit without risking slippage.

Not liking today's entry, however...:mad:
 
Gringotts Bank said:
Not liking today's entry, however...:mad:
Click to expand...

So you bought today at $1.19 and a low as of now is $1.16. This means your discretionary entry has failed to meet your criteria (within 1.5c of low).

Should you simply sell whenever price moves against you by 2c? Would that change your profitability?
 
skc said:
So you bought today at $1.19 and a low as of now is $1.16. This means your discretionary entry has failed to meet your criteria (within 1.5c of low).

Should you simply sell whenever price moves against you by 2c? Would that change your profitability?
Click to expand...

Correct, and that means this trade is highly likely to be a loser (based on history), even though the price has come back to 1.175 now.

A 2c stop loss intraday would change things definitely, but I can't test that without intraday data.

The large, large percentage of profitable trades are going to come when C>O, obviously. But that can happen with an open that gaps lower then finishes higher, even if ROC(c,1) is < 0.
 
Gringotts Bank said:
Correct, and that means this trade is highly likely to be a loser (based on history), even though the price has come back to 1.175 now.

A 2c stop loss intraday would change things definitely, but I can't test that without intraday data.

The large, large percentage of profitable trades are going to come when C>O, obviously. But that can happen with an open that gaps lower then finishes higher, even if ROC(c,1) is < 0.
Click to expand...

I am with skc on this, but would take it logically even further:

If you aren't using a 4 tick stop (based on your criteria) then you aren't actually betting on what you've said you are betting on, rather you're just putting on an intraday, long-only, price action exhaustion strategy without any stops.

My guess is if you use stops to verify your supposed trading hypothesis "I'll be profitable if I can buy LOD+3 ticks", you'll find the results vastly different which essentially means your hypothesis is either wrong or you need to re-make a new hypothesis with inputs from the old one.

As a note, without intraday data to test your actual hypothesis (incl MAE type measurements on intraday and cyclically adjusted basis) you certainly aren't making even an intelligent guess.
 
sinner said:
rather you're just putting on an intraday, long-only, price action exhaustion strategy without any stops.
Click to expand...

That's exactly what I'm doing, except there is a stop. The stop is a time stop (ie. EOD). Where's the problem?

I suggest you read my entry and exit criteria - they're actually very simple.
 
Gringotts Bank said:
That's exactly what I'm doing. The stop is a time stop (ie. EOD). Where's the problem?
Click to expand...

No problem with the execution, just the description.

If you think you are actually buying LOD+3 then every trade should realistically have a entry point-4 stop loss.

On the other hand if "exactly what you're doing" (sic) is buying downside exhaustion, then all that stuff about LOD+3 is just jibber jabber and not even remotely related to your trading hypothesis.

As a note, there is no stop in the form of a contingent order, what you are in fact describing is an "exit". If the market suffers from an unexpected malfunction, or your internet dies a minute before market close or whatever, there is no contingent order or server side exit in place, therefore it's not a stop.

(which means your system should have rules on what to do if you can't exit safely EOD).
 
Gringotts Bank said:
I suggest you read my entry and exit criteria - they're actually very simple.
Click to expand...

Sorry, I have read the whole thread multiple times and have been following it since you started it.

I must be rather daft because I'm still not getting it, other than the broad concept, which I understand well enough.

Can you actually define, algorithmically, what is the entry and exit criteria?
 
sinner said:
No problem with the execution, just the description.

1. If you think you are actually buying LOD+3 then every trade should realistically have a entry point-4 stop loss.

2. On the other hand if "exactly what you're doing" (sic) is buying downside exhaustion, then all that stuff about LOD+3 is just jibber jabber and not even remotely related to your trading hypothesis.

3. As a note, there is no stop in the form of a contingent order, what you are in fact describing is an "exit". If the market suffers from an unexpected malfunction, or your internet dies a minute before market close or whatever, there is no contingent order or server side exit in place, therefore it's not a stop.

(which means your system should have rules on what to do if you can't exit safely EOD).
Click to expand...

1. According to whom? You? I don't HAVE to have a 4 point stop anywhere thanks very much.

2. Buying downside exhaustion. Yep. Well done. LYC offers much greater opportunities for capturing downside exhaustion than any other stock.

3. So now I have to have contingent order too huh? According to whom? You again? Suffice to say if YOUR internet goes down, you might have problems also.

$. You're obviously having a bad day. Go and annoy someone else. None of your complaints is valid.
 
Gringotts Bank said:
1. According to whom? You? I don't HAVE to have a 4 point stop anywhere thanks very much.
Click to expand...

See this is what I don't understand then?

If you buy what you expect is LOD+3, and it goes below your entry point by 4 ticks, obviously your hypothesis is wrong, wouldn't you want to exit?

3. So now I have to have contingent order too huh? According to whom? You again? Suffice to say if YOUR internet goes down, you might have problems also.
Click to expand...

No dude, you don't have to. But it's not really a stop without it. I generally use contingent orders for both stops and take profit if I feel an unexpected technical fault would cause me undue stress. Generally, this means intraday leverage always has stops and tp in place!

$. You're obviously having a bad day. Go and annoy someone else. None of your complaints is valid.
Click to expand...

Jeez dude, no need to jump down my throat, I was just sharing my thoughts. For the record, I'm having a great day.

Here's an example of a range exhaustion model

1. Calculate the daily 20 period average range.
2. Today’s range must expand to exceed #1.
3. The 2-day combined range must expand to exceed 2.2 times of #1.
4. Go long at market if the market is trading in the lower portion of the 2-day range.
5. Protective stop is placed at 0.35 times #1 from the entry price.
6. If not stopped out, exit by 16:00 Eastern Time. i.e. a day trading system
Click to expand...
(h/t NeoTicker)

When I look at that, it seems to have components covering entry, exit and contingencies. Do you see the difference between that simply defined model and your first post? I am still not sure exactly how you define LOD+3?

I posted in this thread because I thought there was some valuable information about exhaustion that could be shared amongst traders, much like the last time I tried to share my thoughts in one of your systems threads. Let me clarify, that other than that, I really don't care how your system works or performs and have no reason to "complain" about your system. Any such perceived "complaints" were solely considered for your benefit.

If you feel I'm annoying you, then I'm happy to no longer try and assist in your threads.
 
sinner said:
See this is what I don't understand then?

1. If you buy what you expect is LOD+3, and it goes below your entry point by 4 ticks, obviously your hypothesis is wrong, wouldn't you want to exit?

2. I am still not sure exactly how you define LOD+3?
Click to expand...

1. A 4 tick stop loss on my discretionary entry might make sense, but I have no way of testing that without intraday data, so I don't know if it would be profitable. What I do know with certainty is that if I can get within 3 ticks of the low, then sell at close each day, then I will make a very high % return. As I have stated this is not a system, it is a discretionary strategy based upon backtests. LYC is a very good stock to try to trade this way.

2. How I define it: Low + 3 ticks = Low + .015 How much more simple could it be?
 
Gringotts Bank said:
2. How I define it: Low + 3 ticks = Low + .015 How much more simple could it be?
Click to expand...

:cautious:

Help me out here GB, I must be really stupid.

How are you actually calculating what you call "Low". Because what I call LOD (Low of Day) is actually something that isn't completely defined until the market closes.

Is what you call "Low" a variable you calculate before the market opens and then you attempt a limit long at around that price, as is the case in the "range exhaustion" model I posted?

Do you wait X seconds/minutes/hours for an intraday low to form and then buy limit long at that price if it goes there again?

Otherwise, how can you possibly define it? At 10:00:01, in hypothetical stock AAA, the OHLC values are identical. O=$1, H=$1, L=$1, C=$1. If AAA then goes down 1 tick, you would still be buying within "3 ticks of the intraday LOD", which has me confused. i.e. given everything you've said, I would still have trouble coding this into a discretionary entry signal.
 
Low is not defined till the EOD. Right.

My backtests are done using EOD data where my entry is defined as L+.015. The tests assume that I will be able to get within 3 ticks of what I anticipate will be the low for the day. A strategy based upon a tested system. Not a system in itself.

If that entry can be achieved, (using my discretionary reading of the market depth), then the % gains will be very very high, much higher than for almost every other stock traded the same way. That's what makes LYC special (at the moment). There's nothing too special about the strategy itself.

The whole idea of this thread is to 1. see if I could do it, and 2. let other people know that if they have good depth reading skills, then this is the stock to trade, and one very good way of trading it.
 
Top